2.34 - Coding of Spectral Information

When sound enters the ear, it is decomposed into its frequency (or “spectral”) components by the auditory periphery, especially the cochlea. The auditory signal is processed at various subcortical levels before reaching the auditory cortex. However, the concept of the cochlea and ascending auditory pathway as a simple frequency analyzer is an incomplete description of the auditory system. This becomes clear at more central levels such as the auditory cortex, where neurons acquire novel response properties necessary to efficiently code complex sound. In the current chapter, we will focus on the primate auditory cortex as a model system. We will first review its computational mechanisms of spectral analysis and then consider more complex mechanisms that facilitate the joint representation of multiple acoustic features of sound, such as pitch or temporal and spectral modulations. We will end on current models of how complex natural sounds are processed to facilitate complex auditory tasks and speech perception.